Hide Forgot
If a libvirt virtual network is configured to support IPv6, it will run an instance of /usr/sbin/radvd to permit IPv6 address discovery on the subnet that is created. Due to the way that radvd is exec'ed (it is daemonized by forking twice, and libvirtd only gets a return code from the first child of these forks, so radvd hasn't been exec'ed yet), no error is logged if radvd isn't installed. Instead everything reports success, the network starts, but address discovery doesn't work. Since many libvirt installations don't use IPv6, it isn't practical to require radvd when installing libvirt. Instead, we need to add a check for executibility of /usr/sbin/radvd prior to the start of the daemonize/exec process. A patch is being submitted upstream as this bug is being filed.
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on therefore solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHBA-2011-0596.html